Abstract: Embodiments are disclosed for training a neural network classifier to learn to more closely align an input image with its attribution map. In particular, in one or more embodiments, the disclosed systems and methods comprise receiving a training image comprising a representation of one or more objects, the training image associated with at least one label for the representation of the one or more objects, generating a perturbed training image based on the training image using a neural network, and training the neural network using the perturbed training image by minimizing a combination of classification loss and attribution loss to learn to align an image with its corresponding attribution map.

Abstract: A method of an electronic device for on-device lifestyle recommendations, includes: receiving a user input; determining a fashion context based on the user input; dynamically clustering fashion objects in at least one image stored in the electronic device based on the fashion context; and displaying a lifestyle recommendation including the clustered fashion objects.

Abstract: A method for enhancing a Quality of Experience (QoE) index, by a wearable device, for a user in an Internet of things (IoT) network comprising a plurality of IoT devices is provided. The method includes monitoring the QoE index of the user in the IoT network, detecting a drop in the QoE index of the user, determining whether the dropped QoE index is less than a QoE threshold, in response to determining that the dropped QoE index is less than the QoE threshold, determining at least one IoT device from the plurality of IoT devices that is responsible for the drop in the QoE index of the user using an Artificial intelligence (AI) model, and controlling the IoT device(s) to raise the QoE index of the user above the QoE threshold.

Abstract: The present disclosure relates to systems, methods, and non-transitory computer readable media for generating trained neural network with increased robustness against adversarial attacks by utilizing a dynamic dropout routine and/or a cyclic learning rate routine. For example, the disclosed systems can determine a dynamic dropout probability distribution associated with neurons of a neural network. The disclosed systems can further drop neurons from a neural network based on the dynamic dropout probability distribution to help neurons learn distinguishable features. In addition, the disclosed systems can utilize a cyclic learning rate routine to force copy weights of a copy neural network away from weights of an original neural network without decreasing prediction accuracy to ensure that the decision boundaries learned are different.

Abstract: Machine-learning based multi-step engagement strategy modification is described. Rather than rely heavily on human involvement to manage content delivery over the course of a campaign, the described learning-based engagement system modifies a multi-step engagement strategy, originally created by an engagement-system user, by leveraging machine-learning models. In particular, these leveraged machine-learning models are trained using data describing user interactions with delivered content as those interactions occur over the course of the campaign. Initially, the learning-based engagement system obtains a multi-step engagement strategy created by an engagement-system user. As the multi-step engagement strategy is deployed, the learning-based engagement system randomly adjusts aspects of the sequence of deliveries for some users.

Abstract: Embodiments disclosed herein describe systems, methods, and products that generate trained neural networks that are robust against adversarial attacks. During a training phase, an illustrative computer may iteratively optimize a loss function that may include a penalty for ill-conditioned weight matrices in addition to a penalty for classification errors. Therefore, after the training phase, the trained neural network may include one or more well-conditioned weight matrices. The one or more well-conditioned weight matrices may minimize the effect of perturbations within an adversarial input thereby increasing the accuracy of classification of the adversarial input. By contrast, conventional training approaches may merely reduce the classification errors using backpropagation, and, as a result, any perturbation in an input is prone to generate a large effect on the output.

Abstract: Methods and systems for customizing the characteristic of an electronic device (in the Internet of Things (IoT) environment based on at least one user's physiological state are provided. The method includes identifying context of the electronic device in response to receiving at least one event by the electronic device, wherein the at least one context includes at least one current user activity and an environmental context of a user. The method includes determining the change in a health parameter of the user and re-calibrates the characteristics of an electronic device through the magnitude of change in health parameter from the learning module. The method includes identifying current user activity and an environment context of the user on receiving the event from the electronic device).

Abstract: Embodiments are disclosed for training a neural network classifier to learn to more closely align an input image with its attribution map. In particular, in one or more embodiments, the disclosed systems and methods comprise receiving a training image comprising a representation of one or more objects, the training image associated with at least one label for the representation of the one or more objects, generating a perturbed training image based on the training image using a neural network, and training the neural network using the perturbed training image by minimizing a combination of classification loss and attribution loss to learn to align an image with its corresponding attribution map.

Abstract: This disclosure includes technologies for image processing, particularly for image generation and editing in a configurable semantic direction. A generative adversarial network is trained with an auxiliary network with an auxiliary task that is designed to disentangle the latent space of the generative adversarial network. Resultantly, a new type of GAN is created to improve image generation or editing in both conditional and unconditional settings.

Abstract: The present disclosure relates to systems, methods, and non-transitory computer readable media for generating trained neural network with increased robustness against adversarial attacks by utilizing a dynamic dropout routine and/or a cyclic learning rate routine. For example, the disclosed systems can determine a dynamic dropout probability distribution associated with neurons of a neural network. The disclosed systems can further drop neurons from a neural network based on the dynamic dropout probability distribution to help neurons learn distinguishable features. In addition, the disclosed systems can utilize a cyclic learning rate routine to force copy weights of a copy neural network away from weights of an original neural network without decreasing prediction accuracy to ensure that the decision boundaries learned are different.

Abstract: Methods and systems for customizing the characteristic of an electronic device (in the Internet of Things (IoT) environment based on at least one user's physiological state are provided. The method includes identifying context of the electronic device in response to receiving at least one event by the electronic device, wherein the at least one context includes at least one current user activity and an environmental context of a user. The method includes determining the change in a health parameter of the user and re-calibrates the characteristics of an electronic device through the magnitude of change in health parameter from the learning module. The method includes identifying current user activity and an environment context of the user on receiving the event from the electronic device).

Abstract: A method for enhancing a Quality of Experience (QoE) index, by a wearable device, for a user in an Internet of things (IoT) network comprising a plurality of IoT devices is provided. The method includes monitoring the QoE index of the user in the IoT network, detecting a drop in the QoE index of the user, determining whether the dropped QoE index is less than a QoE threshold, in response to determining that the dropped QoE index is less than the QoE threshold, determining at least one IoT device from the plurality of IoT devices that is responsible for the drop in the QoE index of the user using an Artificial intelligence (AI) model, and controlling the IoT device(s) to raise the QoE index of the user above the QoE threshold.

Abstract: Methods and systems for customizing the characteristic of an electronic device (in the Internet of Things (IoT) environment based on at least one user's physiological state are provided. The method includes identifying context of the electronic device in response to receiving at least one event by the electronic device, wherein the at least one context includes at least one current user activity and an environmental context of a user. The method includes determining the change in a health parameter of the user and re-calibrates the characteristics of an electronic device through the magnitude of change in health parameter from the learning module. The method includes identifying current user activity and an environment context of the user on receiving the event from the electronic device).

Abstract: Embodiments are disclosed for training a neural network classifier to learn to more closely align an input image with its attribution map. In particular, in one or more embodiments, the disclosed systems and methods comprise receiving a training image comprising a representation of one or more objects, the training image associated with at least one label for the representation of the one or more objects, generating a perturbed training image based on the training image using a neural network, and training the neural network using the perturbed training image by minimizing a combination of classification loss and attribution loss to learn to align an image with its corresponding attribution map.

Abstract: The present disclosure relates to systems, methods, and non-transitory computer readable media for generating trained neural network with increased robustness against adversarial attacks by utilizing a dynamic dropout routine and/or a cyclic learning rate routine. For example, the disclosed systems can determine a dynamic dropout probability distribution associated with neurons of a neural network. The disclosed systems can further drop neurons from a neural network based on the dynamic dropout probability distribution to help neurons learn distinguishable features. In addition, the disclosed systems can utilize a cyclic learning rate routine to force copy weights of a copy neural network away from weights of an original neural network without decreasing prediction accuracy to ensure that the decision boundaries learned are different.

Abstract: A method includes optimizing a video quality of a video data stream received by a user in a video conference. A region of at least one frame of the video data stream may be sampled. The sampling rate may be variable based on the rate of change of objects in the video data stream. A video quality metric corresponding to the video quality of the video data stream may be calculated. The video quality of the video data stream may be adjusted based on the video quality metric. Another method includes retrieving a video and detecting one or more moving regions of interest (ROIs). Each of the ROIs is tagged with metadata configured to allow users to interact with the ROI, and the detected ROIs and their corresponding metadata are stored in a file. Based on the file, playback of the video and movement of the ROIs may be synchronized.

Abstract: Embodiments disclosed herein describe systems, methods, and products that generate trained neural networks that are robust against adversarial attacks. During a training phase, an illustrative computer may iteratively optimize a loss function that may include a penalty for ill-conditioned weight matrices in addition to a penalty for classification errors. Therefore, after the training phase, the trained neural network may include one or more well-conditioned weight matrices. The one or more well-conditioned weight matrices may minimize the effect of perturbations within an adversarial input thereby increasing the accuracy of classification of the adversarial input. By contrast, conventional training approaches may merely reduce the classification errors using backpropagation, and, as a result, any perturbation in an input is prone to generate a large effect on the output.

Abstract: Embodiments disclosed herein describe systems, methods, and products that generate trained neural networks that are robust against adversarial attacks. During a training phase, an illustrative computer may iteratively optimize a loss function that may include a penalty for ill-conditioned weight matrices in addition to a penalty for classification errors. Therefore, after the training phase, the trained neural network may include one or more well-conditioned weight matrices. The one or more well-conditioned weight matrices may minimize the effect of perturbations within an adversarial input thereby increasing the accuracy of classification of the adversarial input. By contrast, conventional training approaches may merely reduce the classification errors using backpropagation, and, as a result, any perturbation in an input is prone to generate a large effect on the output.

Abstract: Methods and systems for customizing the characteristic of an electronic device (in the Internet of Things (IoT) environment based on at least one user's physiological state are provided. The method includes identifying context of the electronic device in response to receiving at least one event by the electronic device, wherein the at least one context includes at least one current user activity and an environmental context of a user. The method includes determining the change in a health parameter of the user and re-calibrates the characteristics of an electronic device through the magnitude of change in health parameter from the learning module. The method includes identifying current user activity and an environment context of the user on receiving the event from the electronic device).

Abstract: The present invention provides fire-retarded rigid polyurethane foam comprising the reaction product of polyol and isocyanate foam forming components and a dialkyl phosphorus-containing compound, namely a reactive mono-hydroxyl-functional dialkyl phosphinates, as flame retardant, serving as highly efficient reactive flame retardant in said rigid polyurethane foam.